During the past 10 years, construction was the leading industry of occupational fatalities when compared to other goods producing industries in the US. This is partially attributed to ineffective safety management strategies, specifically lack of automated construction equipment and worker monitoring. Currently, worker safety performance is measured and recorded manually, assessed subjectively, and the resulting performance information is infrequently shared among selected or all project stakeholders. Accurate and emerging remote sensing technology provides critical spatio-temporal data that has the potential to automate and advance the safety monitoring of construction processes. This doctoral research focuses on pro-active safety utilizing radio-frequency location tracking (Ultra Wideband) and real-time three-dimensional (3D) immersive data visualization technologies. The objective of the research is to create a model that can automatically analyze the spatio-temporal data of the main construction resources (personnel, materials, and equipment), and automatically measure, assess, and visualize worker's safety performance. The research scope is limited to human-equipment interaction in a complex construction site layout where proximities among construction resources are omnipresent. In order to advance the understanding of human-equipment proximity issues, extensive data has been collected in various field trials and from projects with multiple scales. Computational algorithms developed in this research process the data to provide spatio-temporal information that is crucial for construction activity monitoring and analysis. Results indicate that worker's safety performance of selected activities can be automatically and objectively measured using the developed model. The major contribution of this research is the creation of a proximity hazards assessment model to automatically analyze spatio-temporal data of construction resources, and measure, evaluate, and visualize their safety performance. This research will significantly contribute to transform safety measures in construction industry, as it can determine and communicate automatically safe and unsafe conditions to various project participants located on the field or remotely.

Collisions between personnel on foot and heavy equipment or materials on a construction site can be characterized as a contact collision. These types of incidents are a common occurrence on a work site. Technology is needed to improve work zone safety by alerting workers that are in danger of collisions pro-actively and in real-time. Developing this technology may assist in collecting previously un-recorded data on "near-misses" (close-calls). An approach is presented in this paper that is based on wireless radio frequency technology to alert workers in real-time when they are in danger. Various experiments are described that have been conducted in order to gain better understanding of the technology's potential, including measuring equipment blind spots and alert (or safety) zones. Blind spots areas are measured for six common construction vehicles to help determine the required (or minimum) alert distance (safety zone) for the equipment. A computer program was developed in-house to automatically calculate the percentage of blind spots on 2-dimensional planes and in the overall 3-dimensional volume. The blind spots results directly indicate the necessary safety zones for the equipment. The proximity device results show that technology demonstrated the capability of collecting important safety data while pro-actively detecting hazardous situations and warning workers and equipment operators during imminent potential hazardous events. Furthermore, the presented research can lead to improve the overall safety performance in construction and elsewhere through improved learning and education by providing relevant information to decision makers at all levels.

published_or_final_version / Politics and Public Administration / Master / Master of Public Administration

Dissertation submitted in fulfilment of the requirements for the degree Master of Technology: Construction Management Department of the Built Environment in the Faculty of Construction Management and Quantity Surveying at the Cape Peninsula University of Technology 2014 / The construction industry contributes significantly to national economic growth and offers substantial opportunities for job creation; however the industry has continually been plagued by workplace accidents. Moreover, employers may not realize the economic magnitude of workplace injury and ill health arising from construction activities. These accidents represent a considerable economic and social burden to employers, employees and to society as a whole. Despite governments and organisations worldwide maintaining an on-going commitment towards establishing a working environment free of injury and disease, a great deal of construction accidents continues to frequent our society. Given the high rate of construction accidents experienced, employers are not entirely mindful of the actual costs of construction accidents, especially when considering the hidden or indirect costs of accidents. Various safety research efforts have attempted to quantify the true costs of worker injuries, however localised systematic information on cost of construction accidents at work is not readily available from administrative statistical data sources, therefore this study was carried out in order to estimate the costs, like lost workdays or lost income, are clearly visible and can readily be expressed in monetary value; for a large part however, economic consequences of accidents are somewhat hidden. Indirect costs following an accident may be disregarded, damage to the company image is difficult to quantify and pricing human suffering and health damage is subject to discussion. Nevertheless, it is possible to get an adequate insight into the costs of accidents and the potential benefits of accident prevention.

The construction workforce in South Africa is one that is ageing. This is a global phenomenon and necessitates research into how the older cohort of the construction work force can be optimally engaged. Optimum worker productivity, high quality products that meet the specifications required, and high levels of occupational safety and health are integral factors in achieving a sustainable workforce. The purpose of the research was to quantify the injury rates among older construction workers as well as to determine the events leading to these injuries, the nature of the injuries and the bodily locations affected. The costs associated with these injuries were investigated to understand whether there were any discernable differences between injuries to older and younger workers. Apart from the literature review, two statistical construction injury databases were analysed. Qualitative questionnaire based interviews were designed to gather information related to older construction workers. Questionnaires were sent to construction site managers to gauge their perceptions of older construction workers. The statistical data was collected from the Western Cape region and was for the period 1998 through 2005 while the interviews and questionnaire data were collected during 2008. xv The potential benefits to industry are a consolidation of injury information relating to older construction workers. This should assist construction managers with developing policies and implementing strategies to prevent or at least minimise injuries and minimise the related costs, with the aim of more effectively utilising their older workers and ultimately achieving a more sustainable construction industry. The study found that older workers sustained less injuries in total compared with younger workers. No discernable variances occurred between younger and older workers when it came to events leading to injuries (causes) and the type/nature of injuries. It was, however, found that for the body parts affected, older workers were more prone to certain injuries. Older workers sustained less severe injuries compared with their younger counterparts but the injuries were more costly. The research findings supported the notion that older workers receive less training than younger workers.

We are seeing a revolution in computing and artificial intelligence; intelligent machines have become ingrained in and improved every aspect of our lives. Despite the increasing number of intelligent devices and breakthroughs in artificial intelligence, we have yet to achieve truly intelligent environments. Audio is one of the most common sensing and actuation modalities used in intelligent devices. In this thesis, we focus on how we can more robustly integrate audio intelligence into a wide array of resource-constrained platforms that enable more intelligent environments. We present systems and methods for adaptive audio filtering that enables us to more robustly embed acoustic intelligence into a wide range of real time and resource-constrained mobile, embedded, and cyber-physical systems that are adaptable to a wide range of different applications, environments, and scenarios. First, we introduce methods for embedding audio intelligence into wearables, like headsets and helmets, to improve pedestrian safety in urban environments by using sound to detect vehicles, localize vehicles, and alert pedestrians well in advance to give them enough time to avoid a collision. We create a segmented architecture and data processing pipeline that partitions computation between embedded front-end platform and the smartphone platform. The embedded front-end hardware platform consists of a microcontroller and commercial-off-the shelf (COTS) components embedded into a headset and samples audio from an array of four MEMS microphones. Our embedded front-end platform computes a series of spatiotemporal features used to localize vehicles: relative delay, relative power, and zero crossing rate. These features are computed in the embedded front-end headset platform and transmitted wirelessly to the smartphone platform because there is not enough bandwidth to transmit more than two channels of raw audio with low latency using standard wireless communication protocols, like Bluetooth Low-Energy. The smartphone platform runs machine learning algorithms to detect vehicles, localize vehicles, and alert pedestrians. To help reduce power consumption, we integrate an application specific integrated circuit into our embedded front-end platform and create a new localization algorithm called angle via polygonal regression (AvPR) that combines the physics of audio waves, the geometry of a microphone array, and a data driven training and calibration process that enables us to estimate the high resolution direction of the vehicle while being robust to noise resulting from movements in the microphone array as we walk the streets. Second, we explore the challenges in adapting our platforms for pedestrian safety to more general and noisier scenarios, namely construction worker safety sounds of nearby power tools and machinery that are orders of magnitude greater than that of a distant vehicle. We introduce an adaptive noise filtering architecture that allows workers to filter out construction tool sounds and reveal low-energy vehicle sounds to better detect them. Our architecture combines the strengths of both the physics of audio waves and data-driven methods to more robustly filter out construction sounds while being able to run on a resource-limited mobile and embedded platform. In our adaptive filtering architecture, we introduce and incorporate a data-driven filtering algorithm, called probabilistic template matching (PTM), that leverages pre-trained statistical models of construction tools to perform content-based filtering. We demonstrate improvements that our adaptive filtering architecture brings to our audio-based urban safety wearable in real construction site scenarios and against state-of-art audio filtering algorithms, while having a minimal impact on the power consumption and latency of the overall system. We also explore how these methods can be used to improve audio privacy and remove privacy-sensitive speech from applications that have no need to detect and analyze speech. Finally, we introduce a common selective audio filtering platform that builds upon our adaptive filtering architecture for a wide range of real-time mobile, embedded, and cyber-physical applications. Our architecture can account for a wide range of different sounds, model types, and signal representations by integrating an algorithm we present called content-informed beamforming (CIBF). CIBF combines traditional beamforming (spatial filtering using the physics of audio waves) with data driven machine learning sound detectors and models that developers may already create for their own applications to enhance and filter out specified sounds and noises. Alternatively, developers can also select sounds and models from a library we provide. We demonstrate how our selective filtering architecture can improve the detection of specific target sounds and filter out noises in a wide range of application scenarios. Additionally, through two case studies, we demonstrate how our selective filtering architecture can easily integrate into and improve the performance of real mobile and embedded applications over existing state-of-art solutions, while having minimal impact on latency and power consumption. Ultimately, this selective filtering architecture enables developers and engineers to more easily embed robust audio intelligence into common objects found around us and resource-constrained systems to create more intelligent environments.